Year: 2023
Type: article
Abstract: In an era where generative artificial intelligence technology is continuously growing, especially large language models that are receiving significant attention for their application in assisting various research endeavors, this paper analyzes the evolution of machine learning up to the inception of large language models. The discussion highlights DARPA's projects in integrating qualitative research with advanced computational methods, specifically through the Next Generation Social Science (NGS2) and Ground Truth (GT) projects. Subsequently, these initiatives have evolved into Computational Social Science and mixed method research, which combines qualitative (QUAL) and quantitative (QUAN) research methods, resulting in more diverse and comprehensive research outcomes. However, while generative artificial intelligence holds great potential, there are chances of errors, commonly referred to as "hallucination". This paper addresses measures to enhance accuracy concerning this aspect. Furthermore, the paper also presents practical examples of utilizing generative artificial intelligence for trend detection and automated scenario creation.
Source: PAAT Journal
Author: Yuenyong, Kan
Institution: National Institute of Development Administration
Language: Thai
Subfield: General Social Sciences
Field: Social Sciences
Domain: Social Sciences
Open Access status: diamond
OpenAlex ID: W7219425465